1. Technical Field
The present invention relates to an image processing apparatus, a projector and an image processing method.
2. Related Art
In order to suppress flickering and degradation in active matrix type liquid crystal display apparatuses and the like, for example, the liquid crystal can be driven while alternately reversing the voltage that is applied to the liquid crystal between a positive voltage and a negative voltage. In this kind of liquid crystal, due to the effects of the parasitic capacitance between data lines (signal lines, data signal lines, source lines and source electrode lines) and pixel electrodes, the parasitic capacitance between each pixel electrode, the leakage current of the signal with respect to adjacent pixel electrodes and the like, the potentials of other pixel electrodes may change as a result of changes in the potentials of driving target pixel electrodes. Therefore, in a case in which a rectangular pattern or the like is displayed, the positive and negative polarities of changes in the potentials of pixel electrodes may differ and there are cases in which so-called vertical crosstalk, in which lines appear in the vertical direction, occurs due to a luminance difference between the surrounding area occurring at the top and bottom of the corresponding rectangle.
As a method of reducing the generation of this kind of crosstalk, JP-A-2005-77508 discloses a method of correcting the image data of each pixel so that the average potentials in a frame of the pixel electrodes is identical to the average potentials in a frame in a case in which it is assumed that changes in the potentials of the pixel electrodes that accompany changes in the potentials of data lines do not occur.
However, in this kind of method of uniform correction, there are cases in which it is not possible to suitably reduce the generation of crosstalk. More specifically, for example, there are cases in which, due to the characteristics of liquid crystal panels and the like, the changes of each gradation of the pixels is not linear and the corresponding changes increase or decrease according to the gradation, and thus it is not possible to sufficiently correct the corresponding changes with uniform correction.